JP7294691B2 - Disk array device, information system, and storage area construction method - Google Patents

Description

The present invention relates to a disk array device, an information system, and a storage area construction method.

As the capacities of HDDs (Hard disk drives) and SSDs (Solid State Drives) increase, there is an increasing number of situations in which small-capacity HDDs/SSDs are replaced with large-capacity HDDs/SSDs for maintenance or the like. When a RAID (Redundant Array of Inexpensive Disks) is constructed with HDD/SSDs before replacement, due to the characteristics of RAID that HDDs/SSDs with the same properties are used, a large-capacity HDD/SSD is replaced with an HDD/SSD before replacement. It should be used according to capacity. This state is shown in FIG. POOL#0 and POOL#n in FIG. 2A are storage pools constructed by RAID. When replacing the HDD/SSD of POOL#0, a disk with the same capacity as the HDD/SSD constituting POOL#0 is required. The same is true for POOL#n. When the capacity of the HDD/SSD before and after replacement remains the same, as in POOL#n in FIG. is larger than before the replacement, there will be an unused area in the HDD/SSD after the replacement.

In Patent Document 1, when an error occurs in any disk of a disk array device constructed with RAID, the data in the area where the error occurred is stored in an unused area in the disk array device, and subsequent errors are prevented. There is a disclosure of a technique for continuing operation by making access to an area that has caused a failure access to this unused area.

JP 2010-267037 A

If there is an unused area in the HDD/SSD, the unused area will be wasted.

SUMMARY OF THE INVENTION Accordingly, it is an object of the present invention to provide a disk array device, an information system, and a storage area construction method that solve the above-described problems.

According to one aspect of the present invention, a disk array device includes spare area construction means for constructing a spare area, which is a redundant storage area, using unused areas of a plurality of storage devices; and alternative disk construction means for constructing a virtual alternative disk area having the same capacity as the used area in the plurality of storage devices , wherein the alternative disk construction means facilitates degeneracy of the storage area to be substituted. constructing the alternative disk area for the storage area that is likely to degenerate based on the index of the degree of shrinkage;

According to one aspect of the present invention, an information system comprises a plurality of storage devices and the disk array device described above.

According to one aspect of the present invention, a storage area construction method is a storage area construction method executed by a computer, in which unused areas of a plurality of storage devices are used to create redundant storage areas. constructing a certain spare area; constructing a virtual alternative disk area having the same capacity as the used area in the plurality of storage devices for the spare area; and constructing the alternative disk area. In the step, the replacement disk area is constructed for the storage area that is likely to be degenerated based on the index of the easiness of degeneration of the storage area to be replaced.

According to the present invention, it is possible to utilize unused areas (areas not included in the RAID configuration) among a plurality of storage devices redundantly configured by RAID.

1 is a block diagram of an information system according to an embodiment; FIG. FIG. 10 is a diagram showing that unused areas are generated by general disk replacement; It is a figure explaining the outline of the utilization method of the unused area|region which concerns on embodiment. FIG. 10 is a diagram illustrating a spare area created in the case of a RAID type with a redundancy of 1 according to the embodiment; FIG. 10 is a diagram illustrating a spare area created in the case of a RAID type with a redundancy of 2 according to the embodiment; 8 is a flowchart illustrating an example of spare area construction processing according to the embodiment; It is a figure which shows an example of the error management information which concerns on embodiment. 8 is a flowchart illustrating an example of error management information update processing according to the embodiment; 7 is a flowchart illustrating an example of spare area selection processing according to the embodiment; FIG. 4 is a first diagram showing an example of incorporating a virtual alternative disk of the embodiment as a spare storage device; FIG. 10 is a second diagram showing an example of incorporating a virtual alternative disk of the embodiment as a spare storage device; FIG. 11 is a diagram showing an operation at the time of recovery in the configuration of FIG. 10; FIG. 12 is a diagram showing an operation at the time of restoration in the configuration of FIG. 11; 1 is a block diagram showing the configuration of a disk array device having a minimum configuration; FIG. 4 is a flow chart showing processing of a disk array device having a minimum configuration;

<Embodiment>
A disk array device according to an embodiment of the present invention will be described below with reference to the drawings. In the drawings used for the following explanation, the configuration of parts that are not related to the present invention may be omitted and not shown.

(composition)
FIG. 1 is a block diagram of an information system according to an embodiment.
As shown in FIG. 1, an information system 100 comprises a host device 1, a disk array device 2, and a plurality of disks 8-1 to 8-n. The disk 8-1 and the like may be simply referred to as the disk 8 when no particular distinction is required. In the information system 100, a plurality of disks 8 are used to configure one or a plurality of logical disks having redundancy.

The host device 1 includes a processor, memory, etc., and commands the disk array device 2 to write various data to the disk 8 (write command) and read various data from the disk 8 (read command).

The disk array device 2 performs various controls for data transfer between the host device 1 and the disk 8 . The disk array device 2 includes a host processing control unit 3, a cache processing control unit 4, a RAID information management unit 5, a cache 6, a disk processing control unit 7, an error information management unit 9, and a spare area management unit 10. And prepare.

The host processing control unit 3 receives commands from the host and executes processing corresponding to the commands. The cache processing control unit 4 manages addresses and the like stored in the cache 6 and controls input/output of data to/from the cache 6 . The RAID information management unit 5 manages information about logical disks configured using a plurality of disks 8-0 to 8-n. For example, the RAID information management unit 5 manages which address of which disk 8 a command requested by the host device 1 corresponds to. Also, for example, the RAID information management unit 5 performs recovery when the disk 8 is degraded. The cache 6 is a temporary data storage area. A disk processing controller 7 controls data transfer between the cache 6 and the disk 8 . Also, when an error occurs in the disk 8, the disk processing control unit 7 notifies the error information management unit 9 of the error response. The disks 8-0 to 8-n are storage devices such as HDDs or SDDs. A plurality of the disks 8-1 to 8-n are combined to construct one or a plurality of storage areas by RAID technology. Storage areas are classified into an area called a pool (POOL) where user data is stored and a spare area constructed using unused areas of each disk.

For example, when a write command is output from the host device 1 , in the disk array device 2 , the host processing control unit 3 stores data to be written in the cache 6 and reports the completion to the host device 1 . Also, when a read command is received, the RAID information management unit 5 checks which disk 8 in the storage area in which the RAID is constructed corresponds to the data, and outputs the information to the disk processing control unit 7. do. Based on the information obtained from the RAID information management unit 5 , the disk processing control unit 7 reads the data to be read from the appropriate disk 8 to the cache 6 . The host processing control unit 3 transfers the read data to the host device 1 and reports completion.

The error information management section 9 manages error information for each disc 8 . When the disk processing control unit 7 notifies the error information management unit 9 of the error response from the disk 8, the error information management unit 9 updates the number of each error. If it is determined that the error is greater than the threshold, the error information manager 9 instructs the spare area manager 10 via the RAID information manager 5 to allocate a virtual alternative disk.

The spare area management unit 10 constructs a spare area using an unused area of the disk 8, and constructs a virtual alternative disk in the constructed spare area. Now refer to FIG. For example, POOL#0 is constructed by RAID 5, 6, etc. using disks 8-0 to 8-r, and POOL#n is constructed by RIAD 5, 6, etc. using disks 8-m to 8-n. It is also assumed that disks 8-r-2, 8-r-1, and 8-r of POOL#0 have unused areas. The spare area management unit 10 builds one spare area by bundling the unused areas and using the RIADs 5, 6, and the like. A virtual alternative disk can be created in the spare area. For example, if the disk 8-r fails, an area of the same size as the disk 8-r is cut out from the spare area and used as a substitute disk for the disk 8-r. Similarly, for example, if disk 8-n fails, an area of the same size as disk 8-n is cut out from the spare area and used instead of disk 8-n. As a result, the surplus spare area generated by the difference in disk capacity before and after replacement can be used instead of the alternative disk. For example, when the disk 8-r is degraded and needs to be restored, the RAID information management unit 5 inquires of the spare area management unit 10 about an available virtual alternative disk. The spare area management unit 10 allocates a virtual alternative disk to an area having the same size as the size of the disk required for recovery from the spare area. The RAID information management unit 5 generates data to be stored in the disk 8-r from the remaining disks, and restores the data to the virtual alternate disk.

Next, with reference to FIGS. 4 and 5, the process of constructing a virtual alternative disk by the spare area management unit 10 will be described in more detail.
The spare area management unit 10 determines the unused area of each disk, classifies the disks into RAID type disks with a redundancy of 1 (RAID 1 or RAID 5) and disks of a RAID type with a redundancy level of 2 (RAID TM or RAID 6). Build a spare area with RAID type. For example, the spare area management unit 10 constructs a spare area with a RAID type with a redundancy of 1 using an unused area of the disk 8 constructed with a RAID type with a redundancy of 1. FIG. FIG. 4 shows an example of a spare area created for a RAID type with a redundancy of 1. In FIG. POOL#i shown in FIG. 4 is constructed by RAID 5, for example, using five disks 8-0 to 8-4. Each column in FIG. 4 indicates one disk (8-0 to 8-4 are physical disks, 8-4' is a virtual alternate disk). In the range surrounded by POOL#i in FIG. 4, each row indicates a stripe, which is a set of logical data. Furthermore, D01 to D42 are areas (for example, sectors) for storing data in each stripe. P0 to P4 are areas for storing recovery data (parity). S00 to S20 are spare areas. Unused areas exist on the disks 8-2 to 8-4. The spare area management unit 10 constructs a RAID 5 spare area with three or more disks 8-2 to 8-4 having unused areas.

Similarly, the spare area management unit 10 constructs, for example, a RAID 6 spare area using unused areas of the disks 8 constructed with a RAID type having a redundancy of 2. FIG. FIG. 5 shows an example of a spare area created for a RAID type with a redundancy of 2. In FIG. POOL#i shown in FIG. 5 is constructed by RAID 6, for example, using six disks 8-0 to 8-5. D00 to D54 are areas for storing data, and P0 to P4 and Q0 to Q4 are areas for storing restoration data (parity). Unused areas exist on the disks 8-2 to 8-5. The spare area management unit 10 constructs a RAID 6 spare area with four or more disks 8-2 to 8-5 having unused areas.

In addition, the spare area management unit 10 constructs a virtual alternative disk matching the size required for each POOL in the constructed spare area to replace the physical alternative disk. For example, in the example of FIG. 4, the spare area manager 10 constructs a virtual alternate disk 8-4' from the spare area. Then, by using the alternate disk 8-4' as an alternate disk for, for example, the disk 8-4, there is no need for an alternate disk (physical disk) of the same size, and even if the disk 8-4 fails, can be restored. Alternatively, if errors frequently occur in the disk 8-4, the disk 8-4 can be backed up by the alternative disk 8-4' before the failure occurs. Also, the unused areas of the discs 8-2 to 8-4 can be effectively utilized. Similarly, in the example of FIG. 5, the spare area manager 10 constructs a virtual alternate disk 8-5' from the spare area. By doing so, the substitute disk 8-5' can be used, for example, as a substitute disk for the disk 8-5. This eliminates the need to prepare alternative disks of the same size, and allows the unused areas of the disks 8-2 to 8-5 to be effectively utilized. If a virtual alternative disk for the size of each POOL cannot be secured, a spare virtual alternative disk will be created in the order in which the error information management unit 9 determines that the error information management unit 9 is likely to degenerate (in descending order of ease of degeneracy). Allocate space.

In this embodiment, as shown in FIGS. 3 to 5, unused areas of each POOL constructed by RAID with the same redundancy are collected, a spare area is constructed, and a virtual alternative disk is created in the spare area. Build and use as an alternate disk. As a result, the storage area of the disk 8 can be used without waste, and the unit price per capacity can be reduced.

(motion)
1. Virtual Alternative Disk Constructing Process Next, referring to FIG. 6, the process of monitoring error information for each disk 8 and POOL and constructing a virtual alternative disk for a POOL that is likely to degenerate will be described. do. As a premise, it is assumed that the spare area management unit 10 constructs the spare area using unused areas of the disks 8 constructed with the RAID type having the same redundancy.

FIG. 6 is a flowchart illustrating an example of spare area construction processing according to the embodiment.
During operation of the information system 100 , various data are written and read between the host device 1 and the disk 8 . As described above, in the disk array device 2, each functional unit cooperates to control data transfer between the host device 1 and the cache 6, data transfer between the cache 6 and the disk 8, and the like.

While the host device 1 is accessing the disk 8, the error information management unit 9 periodically monitors the error information for each disk 8/POOL based on the notification from the disk processing control unit 7 (step 101). Now refer to FIG. FIG. 7 is a diagram showing an example of error management information. In the example of FIG. 7, POOL#0 consists of disks 8-0, 8-1, . . . , 8-x. The "Check Condition" column for the disk 8-0 in the table of FIG. 7 stores the number of times PC0 an error was detected in the disk 8-0. The "TIMEOUT" column stores the number of times PT0 that a timeout has been detected in the disk 8-0. The "Other Error" column stores the number of times PE0 that other errors have been detected in the disk 8-0. Errors such as "Check Condition", "TIMEOUT", and "Other Errors" are detected by the disk processing control unit 7 and notified to the error information management unit 9. FIG. When the error information management unit 9 receives notification from the disk processing control unit 7 that the "Check Condition" has been detected in the disk 8-0, the error management information PC0 is incremented by one. Updating the error management information will be described later in detail with reference to FIG.

Next, the error information management unit 9 calculates the likelihood of degeneracy (A) for each POOL from the error information for each disk 8/POOL, and identifies POOLs in which degeneracy is likely to occur (step 102). Disk 8/POOL with many failures is highly likely to be degraded. Therefore, POOLs in which degeneracy is likely to occur are specified based on the number of errors. For example, if POOL#0 is composed of disks 8-0 to 8-x, the counts of "Check Condition", "Timeout", and "Other Error" of disk 8-0 are PC0, PT0, and PE0, respectively. , PC1, PT1, and PE1 for disk 8-1, . . . PCx, PTx, and PEx for disk 8-x. , for example, by the following equation (1).

A of POOL#0=WC×(PC0+PC1+...+PCx)+
WT×(PT0+PT1+...+PTx) +
WE×(PE0+PE1+...+PEx)...(1)
Here, WC is a weighting factor for "Check Condition", WT is a weighting factor for "Timeout", and WE is a weighting factor for "Other Errors". The error information management unit 9 also calculates the value of A for the other POOL#1, . Then, for example, the error information management unit 9 identifies the POOL with the largest value of A as the POOL in which degeneracy is likely to occur. Alternatively, the error information management unit 9 identifies POOLs in which the value of A is equal to or greater than a predetermined threshold value as POOLs in which degeneracy is likely to occur.

Next, the error information management unit 9 checks whether the order of the ease of degeneracy (A) for each POOL is changed (step 103). The error information management unit 9 compares the result of step 102 with the order of the ease of degeneracy (A) up to the previous time, and determines whether the order of the ease of degeneracy (A) has changed. If replacement has not occurred (step 103; No), the error information management unit 9 continues monitoring error information. The error information management unit 9 periodically executes the processes after step 102 . For example, if it is the first time after starting to monitor error information for each disk 8/POOL, the determination in step 103 will be Yes. If the determination in step 103 is Yes, a virtual alternative disk is constructed by the following processing.

If replacement occurs (step 103; Yes), the error information management unit 9 sorts the POOL numbers in descending order of easiness of degeneracy (A) (step 104). Next, the spare area management unit 10 checks whether or not a usable virtual alternative disk has already been constructed in descending order of easiness of degeneracy (A) (step 105). The spare area management unit 10 first confirms the POOL with the highest degeneracy (A).

If there is no virtual alternative disk (step 105; Yes), the spare area management unit 10 checks whether there is an empty spare area (step 106). At this time, the spare area management unit 10 confirms whether or not there is a free space in the spare area constructed with the same redundancy RAID type. If there is no free space in the spare area (step 106; No), the spare area management unit 10 issues a release instruction, starting from the POOL with the lowest degeneracy (A) among the spare areas in which the virtual alternative disk is constructed. and create free areas in those spare areas (step 107). At this time, the spare area management unit 10 selects spare areas constructed with a RAID type having the same redundancy as that of the POOL for which the virtual alternative disk is to be constructed, starting from the spare areas with the smallest ease of degeneracy (A) values. Spare areas are released in order. Also, in consideration of an increase in load during degeneration, one or more spare areas constructed in a POOL different from the target POOL may be used to construct a virtual alternative disk. When the spare area management unit 10 creates a free area of the required capacity in the spare area, it uses thin provisioning (only the area is secured and actual writing is not performed) to construct a virtual alternative disk ( step 108).

If a virtual alternative disk has been constructed (step 105; No), or if no virtual alternative disk is necessary (for example, the easiness of degeneration (A) of the target POOL is sufficiently small, etc.), The spare area manager 10 confirms the next POOL (step 109). Specifically, the spare area management unit 10 checks whether or not there is a POOL with the next largest degeneracy (A) value, that is, whether or not all POOLs have been checked. If confirmation of all POOLs has not been completed (step 110; No), the processing from step 105 onwards is executed for the POOL with the next highest degeneracy (A). When confirmation of all POOLs is completed (step 110; Yes), the process from calculation of susceptibility to degeneracy (A) (step 102) is repeated and executed (step 110).

As a result, it is possible to prepare a virtual alternative disk for a POOL that is likely to fail, by utilizing unused areas, without preparing a physical alternative disk. Also, even if a POOL that is prone to disk failure changes during operation of the information system 100, a virtual alternative disk for that POOL can be quickly prepared.

2. Error Information Update Processing Next, the error information update processing will be described using the flowchart of FIG.
FIG. 8 is a flowchart illustrating an example of error management information update processing according to the embodiment.
When the disk processing control unit 7 detects an error during data transfer between the cache 6 and the disk 8, it notifies the error information management unit 9 of the error. Regarding this notification, the error information management unit 9 monitors whether there is an error report (step 120). First, the error information management unit 9 confirms whether or not an error report has been received (step 121). If an error report has been received (step 121; Yes), the error information management unit 9 confirms whether the content of the received error is "Check Condition" (step 122). If it is "Check Condition" (step 122; Yes), the error information management unit 9 updates the "Check Condition counter" of the disk 8 that detected the error (PC0 in FIG. 7 for disk 8-0). (step 123).

If it is not "Check Condition" (step 122; No), the error information management unit 9 confirms whether the content of the received error is "TIMEOUT" (step 124). If it is "TIMEOUT" (step 124; Yes), the error information management unit 9 updates the "TIMEOUT counter" of the disk 8 that detected the error (PT0 in FIG. 7 for disk 8-0) ( step 125). If it is not "TIMEOUT" (step 124; No), the error information management unit 9 stores the "other error counter" (PE0 in FIG. 7 for disk 8-0) of the disk 8 that detected the error. Update (step 126). The error information management unit 9 continues monitoring (step 121) whether or not there is an error report.

If no error has been received (step 121; No), the error information management unit 9 inquires of the RAID information management unit 5 whether or not there is a degraded disk 8 (step 127). If no degeneracy has occurred in any of the disks 8 (step 127; No), the error information management unit 9 continues monitoring whether or not there is an error report (step 121).

If degeneracy has occurred (step 127; Yes), the error information management unit 9 refers to the error management information in FIG. The error information management unit 9 confirms whether or not the most frequently occurring error on the disc 8 is "Check Condition" (step 128). If it is "Check Condition" (step 128; Yes), the error information management unit 9 adds 1 to the "weight of check condition (WC)" (step 129). The error information management unit 9 continues monitoring (step 121) whether or not there is an error report.

If the most frequently occurring error is not "Check Condition" (step 128; No), the error information management unit 9 checks the error management information (FIG. 7) of the degraded disk 8 and Check if the error you are getting is "TIMEOUT" (step 130). If it is "TIMEOUT" (step 130; Yes), the error information management unit 9 adds 1 to the "weight of TIMEOUT (WT)" (step 131). The error information management unit 9 continues monitoring (step 121) whether or not there is an error report. If the most frequently occurring error is not "TIMEOUT" (step 130; No), the error information management unit 9 adds 1 to the "weight of other errors (WE)" (step 132). The error information management unit 9 continues monitoring whether or not there is any error report monitoring.

In this way, by updating the weights WC, WT, and WE according to the error that actually occurs, it is possible to improve the prediction accuracy of the disk 8/POOL in which degradation will occur.

3. Selection of Spare Area Next, a method of selecting a virtual alternative disk when there are a plurality of spare areas will be described with reference to FIG. As a premise, it is assumed that a virtual alternative disk has been constructed in the unused area of the disk 8 by the processing of FIG.
FIG. 9 is a flowchart illustrating an example of spare area selection processing according to the embodiment.
The spare area management unit 10 monitors whether the required virtual alternative disk capacity is sufficient (step 140). First, when the disk 8 is degraded and it is determined that an alternative disk is necessary, the spare area management unit 10 considers that the load will increase and performs Confirm whether or not the constructed virtual alternative disk can be used (step 141). For example, the spare area management unit 10 checks whether there is a free space in the spare area of another POOL constructed with a RAID type having the same redundancy as the POOL in which the disk 8 is degraded. If it is possible to allocate a virtual alternative disk to a spare area constructed in a POOL other than the POOL containing the degraded disk 8 (step 141; Yes), the spare area management unit 10 assigns the spare area , and instructs the RAID information management unit 5 to use the allocated alternative disk for recovery (step 142). The RAID information management unit 5 performs recovery using the assigned alternate disk.

If no virtual alternative disk is assigned to the spare area constructed outside the POOL containing the degraded disk (step 141; No), the spare area management unit 10 determines that the degenerated disk 8 is included. It is checked whether a virtual alternative disk can be assigned to the spare area in the same POOL as the POOL currently being used (step 143). If it is determined that a virtual alternative disk can be allocated to the spare area constructed in the same POOL as the POOL containing the degraded disk (step 143; Yes), the spare area management unit 10 assigns the spare A virtual alternative disk is assigned to the area, and the RAID information management unit 5 is instructed to use the assigned alternative disk for recovery (step 144). The RAID information management unit 5 performs recovery using the assigned alternate disk.

If a virtual alternative disk cannot be assigned to any spare area (step 143; No), the spare area management unit 10 selects the POOL in which the alternative disk exists among the constructed virtual alternative disks. A release instruction is issued in order from the one with the smallest ease of degeneracy (A) value, a necessary free area is created in the spare area, and a virtual alternative disk is constructed using thin provisioning (step 145). Then, the spare area management unit 10 instructs the RAID information management unit 5 to restore using the constructed virtual alternative disk. The RAID information management unit 5 performs recovery using the constructed alternate disk.
As a result, for example, even if a virtual alternative disk that has been constructed in advance by the process of FIG. 6 cannot be used when degeneration occurs, it is possible to quickly allocate a virtual alternative disk and recover the POOL in which the degeneration has occurred. can. In addition, by preferentially allocating a virtual alternative disk of another POOL, it is possible to reduce accesses to the POOL that has degraded during recovery, and reduce the load on the disk.

As described above, according to this embodiment, a spare area is constructed using an unused area when a large-capacity disk 8 (HDD/SSD) is used, and a virtual storage area is created according to the size of each POOL. By constructing such a substitute disk, it becomes possible to use the area of the original disk 8 without waste, and there is no need to separately prepare a substitute disk having the same size as each POOL. As for the spare area, by constructing a RAID with the same redundancy as each POOL (for example, RAID 5 if the redundancy is 1, RAID 6 if the redundancy is 2, etc.), the redundancy is maintained even if the redundancy is degraded. You can keep it from falling. Also, if the size of the spare area is small, such as when the number of large-capacity disks is small, it is not possible to secure virtual alternative disks for all POOLs. By determining the order of priority for allocating virtual alternative disks, it is possible to quickly respond when a virtual alternative disk becomes necessary. It should be noted that the method of constructing a virtual alternative disk according to this embodiment is not limited to the unused area generated by exchanging the disk 8, but can be applied to any unused area of the disk 8 that arises for some reason.

In the above-described embodiment, an example in which the spare area is configured with RAID5 or RAID6 was given, but other RAID types may be used. In the above embodiment, the weights (WC, WT, WE) are updated by adding 1 to the one with the largest number of error occurrences. Or you can update it with another rule. Further, the formula for calculating (A), which is an index of the ease of degeneracy, is not limited to formula (1). For example, the operating time of disk 8 may be taken into consideration. Also, in the above embodiment, an example of constructing a spare area using a free area in the same disk 8 is described, but the spare area may be constructed across a plurality of POOLs with the same redundancy.

<Other embodiments>
Further, in the above-described embodiment, a proposal for constructing a substitute disk in a spare area is described. It may be incorporated as one of a plurality of disks forming a RAID in advance as a storage device. For example, the spare area management unit 10 has the function of building a RAID, and a plurality of disks 8 and a virtual alternative disk may be combined to build one logical disk in which redundancy is compensated. . FIG. 10 shows an example of incorporating a virtual alternative disk of this embodiment into a disk array constructed with RAID5. The left diagram of FIG. 10 shows the construction of a virtual alternate disk 8-5' from the spare area constructed by RAID5. The right diagram of FIG. 10 shows the data storage state after the virtual alternative disk 8-5' is incorporated into the disk 8 constituting the RAID as a spare storage device. For example, in the diagram on the left, data "D00" was written to the pool storage area of disk 8-0, but in the diagram on the right, data "D00" is a sector (*1) of virtual alternative disk 8-5'. (enclosed), and instead the corresponding sector of 8-0 is a spare area ("S00"). The same applies to data "D12", "D31" and "43". Also, parity "P2", which was written in the pool storage area of disk 8-2 in the left figure, is written to the virtual alternate disk 8-5' in the right figure, and the sector of disk 8-2 is a spare area ("S10").

FIG. 11 shows an example of incorporating a virtual alternative disk of this embodiment into a disk array constructed with RAID6. The left diagram of FIG. 11 shows construction of a virtual alternative disk 8-6' from a spare area constructed with RAID 6, and the right diagram incorporates the virtual alternative disk 8-6' as a spare storage device. As a result, the write status of data and parity is widely dispersed. By distributing data writes and distributing the spare area S00 and the like in this way, it is possible to speed up restoration at the time of degeneration. Next, the operation at the time of restoration will be described with reference to FIGS. 12 and 13. FIG.

FIG. 12 shows the operation during recovery in the data write situation of FIG. For example, assume that disk 8-2 is degenerated. For example, for the data in the first row of FIG. 12, the RAID information management unit 5 uses "D00", "D01", "D03", and "P0" to calculate "D02", and stores this value in the spare area. Write to the sector where "S00" is written. Then, after restoration, data is written from the sector to the sector in which "D02" was originally written for restoration. FIG. 13 shows the operation during recovery from the data write situation of FIG. As described in Japanese Patent Laid-Open No. 2018-120382, by distributing and writing data to many disks, it is possible to disperse the spare area S00, etc., and disperse the transfer of data to the spare area that occurs during recovery. and the access load can be reduced. As a result, recovery can be speeded up.

(minimum configuration)
FIG. 14 is a block diagram showing the configuration of a disk array device having a minimum configuration.
The disk array device 200 comprises spare area construction means 201 (spare area management section 10 of the embodiment) and alternative disk construction means 202 (spare area management section 10 of the embodiment).
Spare area construction means 201 constructs a spare area, which is a redundant storage area, using unused areas of a plurality of storage devices. Furthermore, the spare area constructing means 201 may construct the spare area with the same redundancy as a redundant storage area constructed using the used areas of a plurality of storage devices.

The alternative disk constructing means 202 constructs a virtual alternative disk area having the same capacity as the used area in the plurality of storage devices for the spare area. Further, the alternative disk constructing means 202 constructs an alternative disk area for a storage area that is likely to be degenerated (POOL #0, etc.) according to the index (A) of the easiness of degeneracy of the storage area to be replaced. You may In addition, when it is predicted that another storage area is more likely to degenerate than the storage area to be replaced by the already constructed alternative disk area, the alternative disk construction means releases the already constructed alternative disk area. However, the released spare area may be used to construct alternate disk areas for other storage areas (steps 107-108 in FIG. 6, etc.). Also, the alternative disk construction means 202 gives priority to spare areas constructed using unused areas of a plurality of other storage devices over spare areas constructed using unused areas of a plurality of other storage devices. may be used to construct alternate disk areas (steps 141-142 in FIG. 9).

FIG. 15 is a flow chart showing processing of a disk array controller having a minimum configuration.
First, the spare area constructing means 201 constructs a spare area using an unused area generated by, for example, HDD/SSD replacement (step 301). Next, the alternative disk construction means 202 constructs a virtual alternative disk area in the spare area (step 302). The capacity of the virtual alternative disk area may be the same as that of the redundant storage area (disk array of RAID configuration) constructed by multiple HDDs/SSDs. As a result, when a failure or the like occurs in a certain disk, recovery can be performed by allocating a virtual alternative disk area. By using the unused area without waste in this way, the unit price per capacity of the HDD/SSD can be lowered. Also, there is no need to prepare an alternative physical disk.

A part of the disk array devices 2 and 200 in the above-described embodiments may be realized by a computer. In that case, a program for realizing this function may be recorded in a computer-readable recording medium, and the program recorded in this recording medium may be read into a computer system and executed. The "computer system" here is a computer system built in the disk array device 2, 200, and includes hardware such as an OS (Operating System) and peripheral devices.

The term "computer-readable recording medium" refers to portable media such as flexible discs, magneto-optical discs, ROMs and CD-ROMs, and storage devices such as hard discs incorporated in computer systems. Furthermore, "computer-readable recording medium" means a medium that dynamically stores a program for a short period of time, such as a communication line for transmitting a program via a network such as the Internet or a communication line such as a telephone line. It may also include a volatile memory inside a computer system that serves as a server or client in that case, which holds the program for a certain period of time. Further, the program may be for realizing part of the functions described above, or may be capable of realizing the functions described above in combination with a program already recorded in the computer system.

Also, part or all of the disk array devices 2 and 200 in the above-described embodiments may be realized as an integrated circuit such as LSI (Large Scale Integration). Each functional unit of the disk array device 2, 200 may be individually processorized, or part or all may be integrated and processorized. Also, the method of circuit integration is not limited to LSI, but may be realized by a dedicated circuit or a general-purpose processor. In addition, when an integration circuit technology that replaces LSI appears due to advances in semiconductor technology, an integrated circuit based on this technology may be used.

Although one embodiment of the present invention has been described in detail above with reference to the drawings, the specific configuration is not limited to the above-described one, and various design changes and the like can be made without departing from the gist of the present invention. It is possible to Further, one aspect of the present invention can be modified in various ways within the scope of the claims, and an embodiment obtained by appropriately combining technical means disclosed in different embodiments can also be Included in scope. Moreover, it is an element described in each said embodiment and modification, and the structure which replaced the element which shows the same effect is also included.

100... Information system 3... Host processing control unit 4... Cache processing control unit 5... RAID information management unit 6... Cache 7... Disk processing control unit 8... Disk 9. ... error information management section 10 ... spare area management section

Claims (8)

spare area construction means for constructing a spare area, which is a redundant storage area, using unused areas of a plurality of storage devices ;
alternative disk construction means for constructing a virtual alternative disk area having the same capacity as the used area in the plurality of storage devices for the spare area;
with
The alternative disk construction means constructs the alternative disk area targeting the storage area that is likely to degenerate based on an index of the ease of degeneracy of the storage area to be substituted.
Disk array device. The spare area constructing means constructs the spare area with the same redundancy as a redundant storage area configured using the used areas of the plurality of storage devices.
2. The disk array device according to claim 1. The alternative disk constructing means releases the already constructed alternative disk area when it is predicted that other storage areas are more likely to degenerate than the already constructed alternative disk area is to be replaced. , using the released spare area, constructing the alternative disk area with the other storage area as the alternative target;
3. The disk array device according to claim 1 or 2 . further comprising means for monitoring errors occurring in the plurality of storage devices and predicting the likelihood of degeneracy of the storage devices based on the error occurrence status;
The alternative disk building means builds the alternative disk area having the same capacity as the used area based on the used area of the storage device that is likely to degenerate predicted by the predicting means.
4. The disk array device according to any one of claims 1 to 3 . The alternative disk construction means prefers the spare area constructed using unused areas of the plurality of storage devices to the spare area constructed using unused areas of the plurality of storage devices. Constructing the alternative disk area that preferentially uses the used area of the plurality of storage devices as an alternative target;
5. The disk array device according to any one of claims 1 to 4 . Pool storage area construction means for constructing a redundant storage area using the used areas of the plurality of storage devices and the alternative disk area;
6. The disk array device according to any one of claims 1 to 5 , further comprising: a plurality of storage devices;
a disk array device according to any one of claims 1 to 6 ;
Information system with A computer implemented method of constructing a storage area comprising:
constructing a spare area, which is a redundant storage area, using unused areas of a plurality of storage devices ;
constructing a virtual alternative disk area having the same capacity as the used area in the plurality of storage devices for the spare area;
including
In the step of constructing the alternative disk area, constructing the alternative disk area targeting the easily degenerate storage area based on an index of ease of degeneracy of the storage area to be substituted,
How to build the storage area.

Images